June Employment Figures Conflict
By Jeff Pinkerton
Kansas City’s employment shot up in June, adding nearly 15,000 jobs. The labor force also increased by almost 16,000 and the unemployment rate dropped slightly from 6.9 percent to 6.8 percent. This is great news on the local labor front, right?
But wait — the region’s employment dropped in June for the third consecutive month. The employment gains we have experienced over the past year have almost all been erased. We have a net increase of only 3,600 jobs over June 2011, far below what is needed just to absorb new entrants into the workforce, let alone to address the backlog of long-term unemployed.
It may sound like we’re talking about two completely different economies, but both of these storylines about June’s labor market data are true. Confusing? You bet it is. Let’s try to make some sense of it.
First, all of the data we have cited comes from the Bureau of Labor Statistics, but it comes from two separate data series designed to track two different things.
The first dataset, the one with the good news, comes from the Local Area Unemployment Statistics (LAUS). The LAUS is derived from a national household survey and gives us a count of employed people. The LAUS is particularly helpful in giving us a local unemployment rate.
The second dataset, Current Employment Statistics (CES), is a survey of employers that counts jobs, not people. If, for example I lived outside the Kansas City metropolitan statistical area but commuted to a job in the city, I would not be counted in the LAUS data for Kansas City. But I would be counted in the CES data for Kansas City, because my job is there. The CES is generally considered the best source of current, local measure of employment.
Another key difference between the LAUS and the CES is that CES data is seasonally adjusted. Actual employment is prone to seasonal fluctuations. For example, retail businesses often hire more help for the holidays, and vacation-related businesses may hire more in the summer. Seasonally adjusted data is helpful because we can see if an increase in employment signals growing strength in the labor market, or if it is likely from temporary seasonal hiring. Because the LAUS data is not seasonally adjusted at the local level, it makes sense to see the spike in employment that we saw in June.
The chart below compares the LAUS and CES employment data for the Kansas City region from January 2011 through June 2012. The seasonal nature of the LAUS data is evident, as employment peaks in the summer months in both years. The CES data is more consistent and actually is probably more realistic in this case. Using CES, our local data follows national trends. During the first few months of the year we had solid employment growth both regionally and nationwide, but the subsequent employment reports have been disappointing. When adjusted for seasonal fluctuations, the CES shows a decline in employment over the last three months.
The conflicting signals make more sense when you understand the differences in the two series — one is a measure of jobs as opposed to employed people, and one is seasonally adjusted while the other is not.
Still, if employment was trending strongly up or down, we would expect both data series to show it. Different sources of economic data most often conflict when the economy isn’t moving strongly in a particular direction. The very fact that we have such conflicting signals indicates an economy that is still struggling to regain its footing three years after the recession officially ended. The economic recovery has paused, and though our guess is this pause is temporary and that growth will resume more strongly later in the year, such a pause is worrisome. Just how worrisome it is depends on how long it lasts. The next few months will be critical as we move into fall and businesses reassess their hiring plans in preparation for 2013.